Posted by Oghenovo Usiwoma
Dec 14, 2025/20:30 UTC
The email from Novo touches upon an intriguing method of encoding multiple labels within a single 64-bit number by setting the bit positions corresponding to the label numbers to '1'. This concept is introduced with the aim of efficiently representing labels, such as 1, 5, and 10, within one compact number, exemplified by '1058'. However, Novo identifies limitations in this approach, notably its restriction to encoding a maximum of 63 labels due to the bit size, and the resultant difficulty in readability for most individuals. The complexity and impracticality of applying this method to encode thousands of labels into a 64-bit number are acknowledged, deeming it unfeasible.
Further deliberation leads to considering a more streamlined approach whereby only the highest label in a sequence is encoded, predicated on the assumption that labels are incremented by one. By merely encoding the highest label, say '10', the system could infer the presence of all preceding labels up to this maximum. This suggestion aims to simplify the encoding process and enhance comprehensibility, albeit it raises questions about the necessity of initially skipping certain labels, like moving directly from '1' to '5'.
Lastly, while Novo expresses skepticism towards adopting new key formats for descriptors, there's an openness to reevaluate this stance should such formats gain widespread acceptance within the community. This reflects a cautious yet adaptable attitude towards innovations in encoding practices, highlighting the ongoing dialogue within the Bitcoin Development community regarding optimal data representation techniques.
Thread Summary (7 replies)
Dec 4 - Dec 22, 2025
8 messages • 7 replies
TLDR
We’ll email you summaries of the latest discussions from high signal bitcoin sources, like bitcoin-dev, lightning-dev, and Delving Bitcoin.
We'd love to hear your feedback on this project.
Give Feedback